src/test_md5.c

   1 /*
   2 ** 2017-10-13
   3 **
   4 ** The author disclaims copyright to this source code.  In place of
   5 ** a legal notice, here is a blessing:
   6 **
   7 **    May you do good and not evil.
   8 **    May you find forgiveness for yourself and forgive others.
   9 **    May you share freely, never taking more than you give.
  10 **
  11 *************************************************************************
  12 **
  13 ** This file contains code to implement an MD5 extension to TCL.
  14 */
  15 #include "sqlite3.h"
  16 #include <stdlib.h>
  17 #include <string.h>
  18 #include "sqlite3.h"
  19 #include "tclsqlite.h"
  20
  21 /*
  22  * This code implements the MD5 message-digest algorithm.
  23  * The algorithm is due to Ron Rivest.  This code was
  24  * written by Colin Plumb in 1993, no copyright is claimed.
  25  * This code is in the public domain; do with it what you wish.
  26  *
  27  * Equivalent code is available from RSA Data Security, Inc.
  28  * This code has been tested against that, and is equivalent,
  29  * except that you don't need to include two pages of legalese
  30  * with every copy.
  31  *
  32  * To compute the message digest of a chunk of bytes, declare an
  33  * MD5Context structure, pass it to MD5Init, call MD5Update as
  34  * needed on buffers full of bytes, and then call MD5Final, which
  35  * will fill a supplied 16-byte array with the digest.
  36  */
  37
  38 /*
  39  * If compiled on a machine that doesn't have a 32-bit integer,
  40  * you just set "uint32" to the appropriate datatype for an
  41  * unsigned 32-bit integer.  For example:
  42  *
  43  *       cc -Duint32='unsigned long' md5.c
  44  *
  45  */
  46 #ifndef uint32
  47 #  define uint32 unsigned int
  48 #endif
  49
  50 struct MD5Context {
  51   int isInit;
  52   uint32 buf[4];
  53   uint32 bits[2];
  54   unsigned char in[64];
  55 };
  56 typedef struct MD5Context MD5Context;
  57
  58 /*
  59  * Note: this code is harmless on little-endian machines.
  60  */
  61 static void byteReverse (unsigned char *buf, unsigned longs){
  62         uint32 t;
  63         do {
  64                 t = (uint32)((unsigned)buf[3]<<8 | buf[2]) << 16 |
  65                             ((unsigned)buf[1]<<8 | buf[0]);
  66                 *(uint32 *)buf = t;
  67                 buf += 4;
  68         } while (--longs);
  69 }
  70 /* The four core functions - F1 is optimized somewhat */
  71
  72 /* #define F1(x, y, z) (x & y | ~x & z) */
  73 #define F1(x, y, z) (z ^ (x & (y ^ z)))
  74 #define F2(x, y, z) F1(z, x, y)
  75 #define F3(x, y, z) (x ^ y ^ z)
  76 #define F4(x, y, z) (y ^ (x | ~z))
  77
  78 /* This is the central step in the MD5 algorithm. */
  79 #define MD5STEP(f, w, x, y, z, data, s) \
  80         ( w += f(x, y, z) + data,  w = w<<s | w>>(32-s),  w += x )
  81
  82 /*
  83  * The core of the MD5 algorithm, this alters an existing MD5 hash to
  84  * reflect the addition of 16 longwords of new data.  MD5Update blocks
  85  * the data and converts bytes into longwords for this routine.
  86  */
  87 static void MD5Transform(uint32 buf[4], const uint32 in[16]){
  88         register uint32 a, b, c, d;
  89
  90         a = buf[0];
  91         b = buf[1];
  92         c = buf[2];
  93         d = buf[3];
  94
  95         MD5STEP(F1, a, b, c, d, in[ 0]+0xd76aa478,  7);
  96         MD5STEP(F1, d, a, b, c, in[ 1]+0xe8c7b756, 12);
  97         MD5STEP(F1, c, d, a, b, in[ 2]+0x242070db, 17);
  98         MD5STEP(F1, b, c, d, a, in[ 3]+0xc1bdceee, 22);
  99         MD5STEP(F1, a, b, c, d, in[ 4]+0xf57c0faf,  7);
 100         MD5STEP(F1, d, a, b, c, in[ 5]+0x4787c62a, 12);
 101         MD5STEP(F1, c, d, a, b, in[ 6]+0xa8304613, 17);
 102         MD5STEP(F1, b, c, d, a, in[ 7]+0xfd469501, 22);
 103         MD5STEP(F1, a, b, c, d, in[ 8]+0x698098d8,  7);
 104         MD5STEP(F1, d, a, b, c, in[ 9]+0x8b44f7af, 12);
 105         MD5STEP(F1, c, d, a, b, in[10]+0xffff5bb1, 17);
 106         MD5STEP(F1, b, c, d, a, in[11]+0x895cd7be, 22);
 107         MD5STEP(F1, a, b, c, d, in[12]+0x6b901122,  7);
 108         MD5STEP(F1, d, a, b, c, in[13]+0xfd987193, 12);
 109         MD5STEP(F1, c, d, a, b, in[14]+0xa679438e, 17);
 110         MD5STEP(F1, b, c, d, a, in[15]+0x49b40821, 22);
 111
 112         MD5STEP(F2, a, b, c, d, in[ 1]+0xf61e2562,  5);
 113         MD5STEP(F2, d, a, b, c, in[ 6]+0xc040b340,  9);
 114         MD5STEP(F2, c, d, a, b, in[11]+0x265e5a51, 14);
 115         MD5STEP(F2, b, c, d, a, in[ 0]+0xe9b6c7aa, 20);
 116         MD5STEP(F2, a, b, c, d, in[ 5]+0xd62f105d,  5);
 117         MD5STEP(F2, d, a, b, c, in[10]+0x02441453,  9);
 118         MD5STEP(F2, c, d, a, b, in[15]+0xd8a1e681, 14);
 119         MD5STEP(F2, b, c, d, a, in[ 4]+0xe7d3fbc8, 20);
 120         MD5STEP(F2, a, b, c, d, in[ 9]+0x21e1cde6,  5);
 121         MD5STEP(F2, d, a, b, c, in[14]+0xc33707d6,  9);
 122         MD5STEP(F2, c, d, a, b, in[ 3]+0xf4d50d87, 14);
 123         MD5STEP(F2, b, c, d, a, in[ 8]+0x455a14ed, 20);
 124         MD5STEP(F2, a, b, c, d, in[13]+0xa9e3e905,  5);
 125         MD5STEP(F2, d, a, b, c, in[ 2]+0xfcefa3f8,  9);
 126         MD5STEP(F2, c, d, a, b, in[ 7]+0x676f02d9, 14);
 127         MD5STEP(F2, b, c, d, a, in[12]+0x8d2a4c8a, 20);
 128
 129         MD5STEP(F3, a, b, c, d, in[ 5]+0xfffa3942,  4);
 130         MD5STEP(F3, d, a, b, c, in[ 8]+0x8771f681, 11);
 131         MD5STEP(F3, c, d, a, b, in[11]+0x6d9d6122, 16);
 132         MD5STEP(F3, b, c, d, a, in[14]+0xfde5380c, 23);
 133         MD5STEP(F3, a, b, c, d, in[ 1]+0xa4beea44,  4);
 134         MD5STEP(F3, d, a, b, c, in[ 4]+0x4bdecfa9, 11);
 135         MD5STEP(F3, c, d, a, b, in[ 7]+0xf6bb4b60, 16);
 136         MD5STEP(F3, b, c, d, a, in[10]+0xbebfbc70, 23);
 137         MD5STEP(F3, a, b, c, d, in[13]+0x289b7ec6,  4);
 138         MD5STEP(F3, d, a, b, c, in[ 0]+0xeaa127fa, 11);
 139         MD5STEP(F3, c, d, a, b, in[ 3]+0xd4ef3085, 16);
 140         MD5STEP(F3, b, c, d, a, in[ 6]+0x04881d05, 23);
 141         MD5STEP(F3, a, b, c, d, in[ 9]+0xd9d4d039,  4);
 142         MD5STEP(F3, d, a, b, c, in[12]+0xe6db99e5, 11);
 143         MD5STEP(F3, c, d, a, b, in[15]+0x1fa27cf8, 16);
 144         MD5STEP(F3, b, c, d, a, in[ 2]+0xc4ac5665, 23);
 145
 146         MD5STEP(F4, a, b, c, d, in[ 0]+0xf4292244,  6);
 147         MD5STEP(F4, d, a, b, c, in[ 7]+0x432aff97, 10);
 148         MD5STEP(F4, c, d, a, b, in[14]+0xab9423a7, 15);
 149         MD5STEP(F4, b, c, d, a, in[ 5]+0xfc93a039, 21);
 150         MD5STEP(F4, a, b, c, d, in[12]+0x655b59c3,  6);
 151         MD5STEP(F4, d, a, b, c, in[ 3]+0x8f0ccc92, 10);
 152         MD5STEP(F4, c, d, a, b, in[10]+0xffeff47d, 15);
 153         MD5STEP(F4, b, c, d, a, in[ 1]+0x85845dd1, 21);
 154         MD5STEP(F4, a, b, c, d, in[ 8]+0x6fa87e4f,  6);
 155         MD5STEP(F4, d, a, b, c, in[15]+0xfe2ce6e0, 10);
 156         MD5STEP(F4, c, d, a, b, in[ 6]+0xa3014314, 15);
 157         MD5STEP(F4, b, c, d, a, in[13]+0x4e0811a1, 21);
 158         MD5STEP(F4, a, b, c, d, in[ 4]+0xf7537e82,  6);
 159         MD5STEP(F4, d, a, b, c, in[11]+0xbd3af235, 10);
 160         MD5STEP(F4, c, d, a, b, in[ 2]+0x2ad7d2bb, 15);
 161         MD5STEP(F4, b, c, d, a, in[ 9]+0xeb86d391, 21);
 162
 163         buf[0] += a;
 164         buf[1] += b;
 165         buf[2] += c;
 166         buf[3] += d;
 167 }
 168
 169 /*
 170  * Start MD5 accumulation.  Set bit count to 0 and buffer to mysterious
 171  * initialization constants.
 172  */
 173 static void MD5Init(MD5Context *ctx){
 174         ctx->isInit = 1;
 175         ctx->buf[0] = 0x67452301;
 176         ctx->buf[1] = 0xefcdab89;
 177         ctx->buf[2] = 0x98badcfe;
 178         ctx->buf[3] = 0x10325476;
 179         ctx->bits[0] = 0;
 180         ctx->bits[1] = 0;
 181 }
 182
 183 /*
 184  * Update context to reflect the concatenation of another buffer full
 185  * of bytes.
 186  */
 187 static
 188 void MD5Update(MD5Context *ctx, const unsigned char *buf, unsigned int len){
 189         uint32 t;
 190
 191         /* Update bitcount */
 192
 193         t = ctx->bits[0];
 194         if ((ctx->bits[0] = t + ((uint32)len << 3)) < t)
 195                 ctx->bits[1]++; /* Carry from low to high */
 196         ctx->bits[1] += len >> 29;
 197
 198         t = (t >> 3) & 0x3f;    /* Bytes already in shsInfo->data */
 199
 200         /* Handle any leading odd-sized chunks */
 201
 202         if ( t ) {
 203                 unsigned char *p = (unsigned char *)ctx->in + t;
 204
 205                 t = 64-t;
 206                 if (len < t) {
 207                         memcpy(p, buf, len);
 208                         return;
 209                 }
 210                 memcpy(p, buf, t);
 211                 byteReverse(ctx->in, 16);
 212                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 213                 buf += t;
 214                 len -= t;
 215         }
 216
 217         /* Process data in 64-byte chunks */
 218
 219         while (len >= 64) {
 220                 memcpy(ctx->in, buf, 64);
 221                 byteReverse(ctx->in, 16);
 222                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 223                 buf += 64;
 224                 len -= 64;
 225         }
 226
 227         /* Handle any remaining bytes of data. */
 228
 229         memcpy(ctx->in, buf, len);
 230 }
 231
 232 /*
 233  * Final wrapup - pad to 64-byte boundary with the bit pattern
 234  * 1 0* (64-bit count of bits processed, MSB-first)
 235  */
 236 static void MD5Final(unsigned char digest[16], MD5Context *ctx){
 237         unsigned count;
 238         unsigned char *p;
 239
 240         /* Compute number of bytes mod 64 */
 241         count = (ctx->bits[0] >> 3) & 0x3F;
 242
 243         /* Set the first char of padding to 0x80.  This is safe since there is
 244            always at least one byte free */
 245         p = ctx->in + count;
 246         *p++ = 0x80;
 247
 248         /* Bytes of padding needed to make 64 bytes */
 249         count = 64 - 1 - count;
 250
 251         /* Pad out to 56 mod 64 */
 252         if (count < 8) {
 253                 /* Two lots of padding:  Pad the first block to 64 bytes */
 254                 memset(p, 0, count);
 255                 byteReverse(ctx->in, 16);
 256                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 257
 258                 /* Now fill the next block with 56 bytes */
 259                 memset(ctx->in, 0, 56);
 260         } else {
 261                 /* Pad block to 56 bytes */
 262                 memset(p, 0, count-8);
 263         }
 264         byteReverse(ctx->in, 14);
 265
 266         /* Append length in bits and transform */
 267         memcpy(ctx->in + 14*4, ctx->bits, 8);
 268
 269         MD5Transform(ctx->buf, (uint32 *)ctx->in);
 270         byteReverse((unsigned char *)ctx->buf, 4);
 271         memcpy(digest, ctx->buf, 16);
 272 }
 273
 274 /*
 275 ** Convert a 128-bit MD5 digest into a 32-digit base-16 number.
 276 */
 277 static void MD5DigestToBase16(unsigned char *digest, char *zBuf){
 278   static char const zEncode[] = "0123456789abcdef";
 279   int i, j;
 280
 281   for(j=i=0; i<16; i++){
 282     int a = digest[i];
 283     zBuf[j++] = zEncode[(a>>4)&0xf];
 284     zBuf[j++] = zEncode[a & 0xf];
 285   }
 286   zBuf[j] = 0;
 287 }
 288
 289
 290 /*
 291 ** Convert a 128-bit MD5 digest into sequences of eight 5-digit integers
 292 ** each representing 16 bits of the digest and separated from each
 293 ** other by a "-" character.
 294 */
 295 static void MD5DigestToBase10x8(unsigned char digest[16], char zDigest[50]){
 296   int i, j;
 297   unsigned int x;
 298   for(i=j=0; i<16; i+=2){
 299     x = digest[i]*256 + digest[i+1];
 300     if( i>0 ) zDigest[j++] = '-';
 301     sqlite3_snprintf(50-j, &zDigest[j], "%05u", x);
 302     j += 5;
 303   }
 304   zDigest[j] = 0;
 305 }
 306
 307 /*
 308 ** A TCL command for md5.  The argument is the text to be hashed.  The
 309 ** Result is the hash in base64.
 310 */
 311 static int SQLITE_TCLAPI md5_cmd(
 312   void*cd,
 313   Tcl_Interp *interp,
 314   int argc,
 315   const char **argv
 316 ){
 317   MD5Context ctx;
 318   unsigned char digest[16];
 319   char zBuf[50];
 320   void (*converter)(unsigned char*, char*);
 321
 322   if( argc!=2 ){
 323     Tcl_AppendResult(interp,"wrong # args: should be \"", argv[0],
 324         " TEXT\"", (char*)0);
 325     return TCL_ERROR;
 326   }
 327   MD5Init(&ctx);
 328   MD5Update(&ctx, (unsigned char*)argv[1], (unsigned)strlen(argv[1]));
 329   MD5Final(digest, &ctx);
 330   converter = (void(*)(unsigned char*,char*))cd;
 331   converter(digest, zBuf);
 332   Tcl_AppendResult(interp, zBuf, (char*)0);
 333   return TCL_OK;
 334 }
 335
 336 /*
 337 ** A TCL command to take the md5 hash of a file.  The argument is the
 338 ** name of the file.
 339 */
 340 static int SQLITE_TCLAPI md5file_cmd(
 341   void*cd,
 342   Tcl_Interp *interp,
 343   int argc,
 344   const char **argv
 345 ){
 346   FILE *in;
 347   int ofst;
 348   int amt;
 349   MD5Context ctx;
 350   void (*converter)(unsigned char*, char*);
 351   unsigned char digest[16];
 352   char zBuf[10240];
 353
 354   if( argc!=2 && argc!=4 ){
 355     Tcl_AppendResult(interp,"wrong # args: should be \"", argv[0],
 356         " FILENAME [OFFSET AMT]\"", (char*)0);
 357     return TCL_ERROR;
 358   }
 359   if( argc==4 ){
 360     ofst = atoi(argv[2]);
 361     amt = atoi(argv[3]);
 362   }else{
 363     ofst = 0;
 364     amt = 2147483647;
 365   }
 366   in = fopen(argv[1],"rb");
 367   if( in==0 ){
 368     Tcl_AppendResult(interp,"unable to open file \"", argv[1],
 369          "\" for reading", (char*)0);
 370     return TCL_ERROR;
 371   }
 372   fseek(in, ofst, SEEK_SET);
 373   MD5Init(&ctx);
 374   while( amt>0 ){
 375     int n;
 376     n = (int)fread(zBuf, 1, sizeof(zBuf)<=amt ? sizeof(zBuf) : amt, in);
 377     if( n<=0 ) break;
 378     MD5Update(&ctx, (unsigned char*)zBuf, (unsigned)n);
 379     amt -= n;
 380   }
 381   fclose(in);
 382   MD5Final(digest, &ctx);
 383   converter = (void(*)(unsigned char*,char*))cd;
 384   converter(digest, zBuf);
 385   Tcl_AppendResult(interp, zBuf, (char*)0);
 386   return TCL_OK;
 387 }
 388
 389 /*
 390 ** Register the four new TCL commands for generating MD5 checksums
 391 ** with the TCL interpreter.
 392 */
 393 int Md5_Init(Tcl_Interp *interp){
 394   Tcl_CreateCommand(interp, "md5", (Tcl_CmdProc*)md5_cmd,
 395                     MD5DigestToBase16, 0);
 396   Tcl_CreateCommand(interp, "md5-10x8", (Tcl_CmdProc*)md5_cmd,
 397                     MD5DigestToBase10x8, 0);
 398   Tcl_CreateCommand(interp, "md5file", (Tcl_CmdProc*)md5file_cmd,
 399                     MD5DigestToBase16, 0);
 400   Tcl_CreateCommand(interp, "md5file-10x8", (Tcl_CmdProc*)md5file_cmd,
 401                     MD5DigestToBase10x8, 0);
 402   return TCL_OK;
 403 }
 404
 405 /*
 406 ** During testing, the special md5sum() aggregate function is available.
 407 ** inside SQLite.  The following routines implement that function.
 408 */
 409 static void md5step(sqlite3_context *context, int argc, sqlite3_value **argv){
 410   MD5Context *p;
 411   int i;
 412   if( argc<1 ) return;
 413   p = sqlite3_aggregate_context(context, sizeof(*p));
 414   if( p==0 ) return;
 415   if( !p->isInit ){
 416     MD5Init(p);
 417   }
 418   for(i=0; i<argc; i++){
 419     const char *zData = (char*)sqlite3_value_text(argv[i]);
 420     if( zData ){
 421       MD5Update(p, (unsigned char*)zData, (int)strlen(zData));
 422     }
 423   }
 424 }
 425 static void md5finalize(sqlite3_context *context){
 426   MD5Context *p;
 427   unsigned char digest[16];
 428   char zBuf[33];
 429   p = sqlite3_aggregate_context(context, sizeof(*p));
 430   MD5Final(digest,p);
 431   MD5DigestToBase16(digest, zBuf);
 432   sqlite3_result_text(context, zBuf, -1, SQLITE_TRANSIENT);
 433 }
 434 int Md5_Register(
 435   sqlite3 *db,
 436   char **pzErrMsg,
 437   const sqlite3_api_routines *pThunk
 438 ){
 439   int rc = sqlite3_create_function(db, "md5sum", -1, SQLITE_UTF8, 0, 0,
 440                                  md5step, md5finalize);
 441   sqlite3_overload_function(db, "md5sum", -1);  /* To exercise this API */
 442   return rc;
 443 }